Neuron Explorer

Qwen3-0.6B

unclear: weak math-topic words

A neuron in layer 4 of Qwen3-0.6B. Its strongest snippets are mixed: they peak on different tokens. It is featured, so it has steering outputs.

What this neuron does

This neuron responds only weakly, mostly to technical words in mathematics text such as 'singular' and 'inversion'. Its pattern is faint and mixed, so it is hard to describe simply.

Written by Claude from this neuron's activations and steering outputs. Confidence: low. Think it is wrong? Flag it (opens a prefilled issue on GitHub).

Fires most on

The text windows where this neuron's activation was highest. Amber marks positive activation; the outlined token is the peak. Hover a token for its strength.

  1. #1 peak

    The study of the real points of an algebraic variety is the subject of real algebraic geometry.⏎- A large part of singularity theory is devoted to the singularities of algebraic varieties.⏎- When computers became more common, a field named

  2. #2 peak

    plane belongs to an algebraic curve if its coordinates satisfy a given polynomial equation. Basic questions involve the study of the points of special interest like the singular points, the inflection points and the points at infinity. More advanced questions involve

  3. #3 peak

    not insignificant. The author compares these odds to other unlikely events and includes ideas on how the government could help lower these odds.⏎%0 Electronic Source %A Plait, Phil %D December 6, 2010 %T Bad

  4. #4 peak

    ID 5 volumes cannot be striped, concatenated, or mirrored.⏎Do not create a RAID 5 volume from a component that contains an existing file system. Doing so will erase the data during the RAID 5 initialization process.⏎A RAID 5

  5. #5 peak

    and even special effects in movies. It is essentially the evolution of a merger between a graphical user interface and Non-Uniform Rational B-Splines (NURBS) geometry, and currently functions for all major platforms.⏎Become

  6. #6 peak

    ordinary and the other is an antiparticle. With respect to symmetries and invariance principles, three special conservation laws have been described, associated with inversion or reversal of space, time, and charge.⏎Conservation laws are considered to

Steering

Push this neuron up or down while the model writes, and watch what changes. The outputs were generated ahead of time, so nothing runs in your browser.

turned down a lot
Prompt

Here is something I have been thinking about lately: the number of people who are in a group. The number of people who are in a group is 200. Now, the number of people in a group is 100. The number of people in a group is

Most negative on

Where the activation went furthest below zero. Blue marks negative activation. Many neurons mean something different in each direction.

  1. #1 peak

    a branch of mathematics studying zeros of polynomial equations. Modern algebraic geometry is based on more abstract techniques of abstract algebra, especially commutative algebra, with the language and the problems of geometry.⏎The main objects of

  2. #2 peak

    CC BY-NC-SA⏎Given an integer , the exponential sum version of the Gauss sum modulo is defined by⏎and the quadratic residue version is defined by⏎where is the Jacobi symbol, which has values

  3. #3 peak

    representability of via the Yoneda lemma. That is, any element induces, in natural bijective fashion, a natural transformation⏎and is universal if is an isomorphism.⏎Thus, universal elements are part and parcel of any discussion involving

  4. #4 peak

    both sides. Part A: Short Answer (1 problems, 12 points). 1. Let S and T be linear spaces and A : S T be a linear map. Say V and W are particular solutions of the equations A V

Connections

Show on map

Which neurons in earlier layers feed this one, and which later ones it feeds. Amber links excite, blue links suppress; longer bars are stronger.

Fed by earlier layers

  1. L3:25 suppresses
  2. L3:461 excites
  3. L3:924 suppresses
  4. L3:52 suppresses
  5. L3:105 excites
  6. L0:2536 weakly excites
  7. L2:84 weakly excites
  8. L3:185 weakly suppresses
  9. L3:2272 weakly suppresses
  10. L3:1258 weakly excites
  11. L2:454 weakly excites
  12. L3:314 weakly excites
  13. L2:122 weakly excites
  14. L3:408 weakly excites
  15. L0:282 weakly suppresses
  16. L3:278 weakly excites
  17. L1:3029 weakly excites
  18. L1:2387 weakly excites
  19. L0:1955 weakly suppresses
  20. L3:703 weakly suppresses
  21. L1:307 weakly suppresses
  22. L3:421 weakly suppresses
  23. L0:1495 weakly suppresses
  24. L0:1475 weakly excites

Feeds later layers

  1. L8:312 weakly suppresses
  2. L8:383 weakly excites
  3. L5:2180 weakly excites
  4. L9:2326 weakly excites
  5. L8:3028 weakly suppresses
  6. L12:1870 weakly excites
  7. L9:932 weakly suppresses
  8. L10:659 weakly suppresses
  9. L16:1656 weakly excites
  10. L15:2913 weakly excites
  11. L20:1031 weakly excites
  12. L5:1128 weakly suppresses
  13. L9:878 weakly excites
  14. L12:2315 weakly excites
  15. L6:1371 weakly excites
  16. L15:873 weakly suppresses
  17. L11:2975 weakly excites
  18. L12:1341 weakly suppresses
  19. L18:2125 weakly suppresses
  20. L18:2437 weakly excites
  21. L22:1365 weakly excites
  22. L23:1645 weakly excites
  23. L7:1989 weakly excites
  24. L15:2561 weakly suppresses

Activation histogram

How often this neuron fires at each strength. Almost always it does close to nothing (the tall bars in the middle); the snippets above are the rare moments it fired hard, amber for positive and blue for negative.

1 values between -0.62 and -0.591 values between -0.59 and -0.551 values between -0.49 and -0.461 values between -0.46 and -0.424 values between -0.42 and -0.397 values between -0.39 and -0.3612 values between -0.36 and -0.338 values between -0.33 and -0.2921 values between -0.29 and -0.2643 values between -0.26 and -0.2357 values between -0.23 and -0.2094 values between -0.20 and -0.16106 values between -0.16 and -0.13180 values between -0.13 and -0.10262 values between -0.10 and -0.07303 values between -0.07 and -0.03435 values between -0.03 and -0.00447 values between -0.00 and 0.03388 values between 0.03 and 0.07376 values between 0.07 and 0.10321 values between 0.10 and 0.13243 values between 0.13 and 0.16180 values between 0.16 and 0.20147 values between 0.20 and 0.23104 values between 0.23 and 0.2660 values between 0.26 and 0.2941 values between 0.29 and 0.3326 values between 0.33 and 0.3619 values between 0.36 and 0.3913 values between 0.39 and 0.424 values between 0.42 and 0.461 values between 0.46 and 0.491 values between 0.52 and 0.551 values between 0.65 and 0.68offnegativepositiveoften

Logit lens

The words this neuron nudges the model toward (amber) or away from (blue) when it fires. A hint, not a label: many neurons work through later layers instead.

Promotes

  1. TRACT
  2. 形式
  3. 大海
  4. SEG
  5. 树枝
  6. iona
  7. NONE
  8. 版本
  9. 的变化
  10. 效果

Suppresses

  1. her
  2. iland
  3. ivement
  4. .hibernate
  5. iven
  6. ernen
  7. initialised
  8. tipped
  9. PropertyInfo
  10. ement